Overall Goal: Our long-term goal is to gain a molecular understanding of the machinery that regulates the release of infectious HIV virions from infected cells. In our parent RO1 award which started on 9/1/2018 we proposed to establish the localization and kinetics of Gag-Pol incorporation in HIV virions using iPALM microscopy. This request for capital funds is for purchase of 4 computer nodes to facilitate the data processing of iPALM. In our RO1 application, we specifically proposed to: ?AIM2-b) Localization of Gag-Pol within immature HIV lattice using correlative light and electron microscopy: Initiation of Gag-Pol auto-processing requires proximity of at least two Gag-Pol molecules within the immature HIV lattice. Therefore, understanding the structural distribution of Gag-Pol within the HIV lattice is critical for our mechanistic understanding of auto-processing initiation. Interference based Photoactivated Localization Microscopy (iPALM) has 10 nm axial and 20 nm in-plane resolution, respectively. We have visualized single Gag-Dendra proteins within the lattice of immature VLPs by correlating iPALM and SEM imaging on purified Gag VLPs. Here we propose to use this methodology to measure the distribution of Gag-Pol:Gag-Pol distances within immature HIV virions, this distribution will be used to discriminate between various models of Gag-Pol auto-processing.? Progress on infrastructure: To achieve this AIM we have purchased an iPALM microscope which has been installed in the PI?s lab at the Crocker Science Building in May 2018 (see MOU attached). The microscopy lab is located in the basement of the new Crocker Science Building which is specially built for reducing vibration and temperature fluctuations. Since the instrument has been installed we have measured the vibration and temperature stability in the room and verified that we can maintain the drift within a window of 25 nm/hr. This specific drift threshold is very encouraging since it will allow extending the total measurement window to a few hours which in turn will allow localization of every Dendra molecule incorporated in each virion. Since there are in average only 100 Pol molecules within each HIV virion, localizing each one is of essential importance and therefore long experimental runs ensuring activation of every Dendra molecule associated with Pol is essential. Technical complications: The iPALM scope generates 3 full frame images at each exposure, a typical experiment generates between 30,000 to 300,000 images. Each image needs to be analyzed using a Gaussian fitting function to identify single molecules. Analyzing 30,000 images which are acquired within 30 minutes on the scope using a powerful 12 core PC can take between 6-10 hours, therefore data processing is an essential bottleneck in the experiments. Solutions: To overcome the technical difficulty of analyzing the iPALM, Harold Hess lab at the HHMI Janelia campus has developed a version of the analysis software that runs on Linux clusters (PeakSelector). During Fall of 2017, PI and one of his graduate students along with a staff member of the University of Utah?s Center for High Performance Computing (CHPC) visited HHMI Janelia campus and successfully copied the full analysis code onto the CHPC servers at Utah. Without the purchase of nodes however we would not be able to effectively use the CHPC resources. Testing: In June of 2018, we tested a data set of 30,000 frames using a borrowed 4 node cluster. We were able to analyze the data within 15 min, which is an amazing improvement. Therefore we are confident if we can obtain the 4 nodes as outlined in the Lenovo Quote (see attached) and supported by CHPC as specified in the agreement between Saffarian and CHPC (see attached), we would solve our analysis problem.